We will be exploring the possibility that consistent alterations of DNA methylation and endocrine differentiation, found in established human lung cancers, may mark the earliest stages of development of these tumors. The goal is to develop new biomarkers for early detection and for understanding events in the genesis of lung neoplasms. Using a strategy inherent to Projects 1 and 3 in this SPORE proposal, we will first explore these changes in isolated bronchial epithelium and established tumors from patients with each of the major histologic types of lung cancer. In collaboration with Project 4, we will also study similar samples from uranium miners. Positive findings will be pursued in a second series of studies to evaluate marker efficacy in sputum and bronchoscopy specimens of patients at risk for developing lung cancer. In parallel studies, DNA methylation patterns are also being explored in cultured human bronchial epithelial cells into which genes altered in lung cancers have been inserted. For DNA methylation, we will focus on areas of regional hypermethylation which have been found, early, in human tumor progression. These changes have potential for abnormally silencing, or marking chromatin changes which silence, expression of genes on chromosomes consistently altered in lung cancer. Increases in expression of the DNA-methyltransferase (DNA-MT) gene, which also occur early in colon cancer progression, will also be sought. For endocrine differentiation, parameters which characterize early cellular responses of hamster tracheal epithelium to carcinogens, and which cluster with the human small cell lung cancer phenotype, will be studied. These markers may not only be useful for early diagnosis of lung cancer, in general, but also specifically for SCLC.